py

Name: py
Author: TencentCloudBase

25stars1forks

How to install py?

tencentcloudbase/cloudbase-skills/py

$npx -y skills add TencentCloudBase/cloudbase-skills --skill py

Installs into the current project.

Summary

Build production-ready AI agent backends using the CloudBase Agent Python SDK — create agents with LangGraph/CrewAI/LlamaIndex, serve them via FastAPI with AG-UI protocol streaming + OpenAI-compatible endpoints, add tools (bash, filesystem, MCP, code execution), memory (in-memory

Files · 1

View on GitHub

SKILL.md

1	# CloudBase Agent Python SDK
2
3	Build production-ready AI agent backends with multi-framework support, streaming
4	protocol, rich tools, persistent memory, and full observability.
5
6	> Note: This skill is for Python projects only.
7
8	## When to use this skill
9
10	Use this skill for AI agent development when you need to:
11
12	- Deploy AI agents as HTTP services with AG-UI protocol support
13	- Build agent backends using LangGraph, CrewAI, or LlamaIndex frameworks
14	- Create custom agent adapters implementing the AbstractAgent interface
15	- Understand AG-UI protocol events and message streaming
16	- Build production-ready agent servers with FastAPI
17
18	Do NOT use for:
19	- Simple AI model calling without agent capabilities (use `ai-model-*` skills)
20	- CloudBase cloud functions (use `cloud-functions` skill)
21	- CloudRun backend services without agent features (use `cloudrun-development` skill)
22	- TypeScript/JavaScript agent projects (use `cloudbase-agent` skill, refer to the `ts/` sub-directory)
23
24	## How to use this skill (for a coding agent)
25
26	1. Choose the right adapter
27	- Use LangGraph adapter for stateful, graph-based workflows
28	- Use CrewAI adapter for multi-agent collaboration patterns
29	- Build custom adapter for specialized agent logic
30
31	2. Write agent code — follow the adapter-specific doc from the Routing table
32
33	3. Deploy the agent server — follow the blocking deployment pipeline in [agent-deployment](agent-deployment.md)
34
35	## Routing (Execution Order)
36
37	> ⚠️ Deployment is a BLOCKING 4-step pipeline. Steps marked ✅ BLOCKING
38	> must be completed AND verified before proceeding to the next step.
39	> Do NOT call `manageAgent` until all blocking steps pass.
40
41	\| Step \| Task \| Document \| Blocking? \|
42	\|------\|------\|----------\|-----------\|
43	\| 0 \| Choose adapter & write agent code \| See "Adapter Selection" below \| — \|
44	\| 1 \| Ensure Python 3.10 \| [agent-deployment](agent-deployment.md) § Step 1 \| ✅ BLOCKING \|
45	\| 2 \| Build env/ (one-shot) \| [agent-deployment](agent-deployment.md) § Step 2 \| ✅ BLOCKING \|
46	\| 3 \| Verify env/ integrity \| [agent-deployment](agent-deployment.md) § Step 3 \| ✅ BLOCKING \|
47	\| 4 \| Deploy with manageAgent \| [agent-deployment](agent-deployment.md) § Step 4 \| — \|
48
49	### Adapter Selection (Step 0)
50
51	\| Framework \| Read \| Install \|
52	\|-----------\|------\|---------\|
53	\| LangGraph (stateful graphs) \| [adapter-langgraph](adapter-langgraph.md) \| `cloudbase-agent-langgraph` \|
54	\| CrewAI (multi-agent crews) \| [adapter-development](adapter-development.md) \| `cloudbase-agent-crewai` \|
55	\| Coze platform \| [adapter-coze](adapter-coze.md) \| `cloudbase-agent-coze` \|
56	\| Custom / raw FastAPI \| [server-quickstart](server-quickstart.md) + [adapter-development](adapter-development.md) \| `cloudbase-agent-server` \|
57
58	### Additional References (read on demand, NOT required for deployment)
59
60	\| Task \| Read \|
61	\|------\|------\|
62	\| Server setup, middleware, multi-agent, CORS \| [server-quickstart](server-quickstart.md) \|
63	\| Authentication and user context \| [authentication](authentication.md) \|
64
65	## Quick Start (Framework-Agnostic)
66
67	Prerequisites: Python >= 3.10 is required.
68
69	1. Install dependencies (pick ONE adapter):
70
71	```bash
72	# Option A: LangGraph-based agent
73	pip install cloudbase-agent-langgraph
74
75	# Option B: CrewAI-based agent
76	pip install cloudbase-agent-crewai
77
78	# Option C: Custom / minimal
79	pip install cloudbase-agent-server
80	```
81
82	2. Create server entry point:
83
84	```python
85	# server.py — this pattern works with ANY adapter
86	import os
87	from dotenv import load_dotenv
88	load_dotenv()
89
90	from cloudbase_agent.server import AgentServiceApp, AgentCreatorResult
91
92	# Import your agent (framework-specific, see adapter docs)
93	# from agents.chat.agent import create_my_agent
94
95	def create_agent() -> AgentCreatorResult:
96	agent = create_my_agent() # Your agent factory
97	return {"agent": agent}
98
99	app = AgentServiceApp()
100	app.set_cors_config(allow_origins=["*"])
101
102	if __name__ == "__main__":
103	port = int(os.environ.get("SCF_RUNTIME_PORT", "9000"))
104	app.run(create_agent, port=port, host="0.0.0.0")
105	```
106
107	3. Deploy to CloudBase:
108
109	Follow the 4-step deployment pipeline in [agent-deployment](agent-deployment.md).
110
111	---
112
113	## Architecture
114
115	```
116	Client (React / MiniProgram / curl)
117	│ HTTP POST + SSE streaming
118	▼
119	┌─────────────────────

People who liked this also viewed

6 picks

Type

py

25stars1forks

View on GitHub

How to install py?

tencentcloudbase/cloudbase-skills/py

$npx -y skills add TencentCloudBase/cloudbase-skills --skill py

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# CloudBase Agent Python SDK
2
3	Build production-ready AI agent backends with multi-framework support, streaming
4	protocol, rich tools, persistent memory, and full observability.
5
6	> Note: This skill is for Python projects only.
7
8	## When to use this skill
9
10	Use this skill for AI agent development when you need to:
11
12	- Deploy AI agents as HTTP services with AG-UI protocol support
13	- Build agent backends using LangGraph, CrewAI, or LlamaIndex frameworks
14	- Create custom agent adapters implementing the AbstractAgent interface
15	- Understand AG-UI protocol events and message streaming
16	- Build production-ready agent servers with FastAPI
17
18	Do NOT use for:
19	- Simple AI model calling without agent capabilities (use `ai-model-*` skills)
20	- CloudBase cloud functions (use `cloud-functions` skill)
21	- CloudRun backend services without agent features (use `cloudrun-development` skill)
22	- TypeScript/JavaScript agent projects (use `cloudbase-agent` skill, refer to the `ts/` sub-directory)
23
24	## How to use this skill (for a coding agent)
25
26	1. Choose the right adapter
27	- Use LangGraph adapter for stateful, graph-based workflows
28	- Use CrewAI adapter for multi-agent collaboration patterns
29	- Build custom adapter for specialized agent logic
30
31	2. Write agent code — follow the adapter-specific doc from the Routing table
32
33	3. Deploy the agent server — follow the blocking deployment pipeline in [agent-deployment](agent-deployment.md)
34
35	## Routing (Execution Order)
36
37	> ⚠️ Deployment is a BLOCKING 4-step pipeline. Steps marked ✅ BLOCKING
38	> must be completed AND verified before proceeding to the next step.
39	> Do NOT call `manageAgent` until all blocking steps pass.
40
41	\| Step \| Task \| Document \| Blocking? \|
42	\|------\|------\|----------\|-----------\|
43	\| 0 \| Choose adapter & write agent code \| See "Adapter Selection" below \| — \|
44	\| 1 \| Ensure Python 3.10 \| [agent-deployment](agent-deployment.md) § Step 1 \| ✅ BLOCKING \|
45	\| 2 \| Build env/ (one-shot) \| [agent-deployment](agent-deployment.md) § Step 2 \| ✅ BLOCKING \|
46	\| 3 \| Verify env/ integrity \| [agent-deployment](agent-deployment.md) § Step 3 \| ✅ BLOCKING \|
47	\| 4 \| Deploy with manageAgent \| [agent-deployment](agent-deployment.md) § Step 4 \| — \|
48
49	### Adapter Selection (Step 0)
50
51	\| Framework \| Read \| Install \|
52	\|-----------\|------\|---------\|
53	\| LangGraph (stateful graphs) \| [adapter-langgraph](adapter-langgraph.md) \| `cloudbase-agent-langgraph` \|
54	\| CrewAI (multi-agent crews) \| [adapter-development](adapter-development.md) \| `cloudbase-agent-crewai` \|
55	\| Coze platform \| [adapter-coze](adapter-coze.md) \| `cloudbase-agent-coze` \|
56	\| Custom / raw FastAPI \| [server-quickstart](server-quickstart.md) + [adapter-development](adapter-development.md) \| `cloudbase-agent-server` \|
57
58	### Additional References (read on demand, NOT required for deployment)
59
60	\| Task \| Read \|
61	\|------\|------\|
62	\| Server setup, middleware, multi-agent, CORS \| [server-quickstart](server-quickstart.md) \|
63	\| Authentication and user context \| [authentication](authentication.md) \|
64
65	## Quick Start (Framework-Agnostic)
66
67	Prerequisites: Python >= 3.10 is required.
68
69	1. Install dependencies (pick ONE adapter):
70
71	```bash
72	# Option A: LangGraph-based agent
73	pip install cloudbase-agent-langgraph
74
75	# Option B: CrewAI-based agent
76	pip install cloudbase-agent-crewai
77
78	# Option C: Custom / minimal
79	pip install cloudbase-agent-server
80	```
81
82	2. Create server entry point:
83
84	```python
85	# server.py — this pattern works with ANY adapter
86	import os
87	from dotenv import load_dotenv
88	load_dotenv()
89
90	from cloudbase_agent.server import AgentServiceApp, AgentCreatorResult
91
92	# Import your agent (framework-specific, see adapter docs)
93	# from agents.chat.agent import create_my_agent
94
95	def create_agent() -> AgentCreatorResult:
96	agent = create_my_agent() # Your agent factory
97	return {"agent": agent}
98
99	app = AgentServiceApp()
100	app.set_cors_config(allow_origins=["*"])
101
102	if __name__ == "__main__":
103	port = int(os.environ.get("SCF_RUNTIME_PORT", "9000"))
104	app.run(create_agent, port=port, host="0.0.0.0")
105	```
106
107	3. Deploy to CloudBase:
108
109	Follow the 4-step deployment pipeline in [agent-deployment](agent-deployment.md).
110
111	---
112
113	## Architecture
114
115	```
116	Client (React / MiniProgram / curl)
117	│ HTTP POST + SSE streaming
118	▼
119	┌─────────────────────

People who liked this also viewed

6 picks

Type